This Phase-II proposal builds on the work of Phase I, together with interim work funded by the New York State Science and Technology Foundation, in further developing and strengthening a computer system aimed at (a) providing information about psychopharma- cologicals, (b) providing regularly scheduled reinforcement and monitoring of drug compliance, (c) collecting self-report side effects data, (d) providing illness severity information to therapists about patients with affective disorders. The second aim of the project is to evaluate, using a quasi-experimental design, specified outcomes associated with use of the system (blood serum levels, pill counts, morbid events, patient & therapist satisfaction). Clinic psychiatric outpatients with one of the DSM- III diagnoses (major depression, bipolar disorder:manic, dysthymia) and who are beginning a psychopharmacological drug regimen will be randomly assigned to experimental (computer intervention) or to control groups (one placebo, one no placebo). Patients will be assessed on a number of factors (possible covariates) in addition to outcomes as a means to examining interrelationships among factors predicting compliance and side effect outcomes. The experimental design hypotheses will be tested using logistic regression and other analytic techniques. The system being developed and field tested is intended for use in clinics or in private practice groups as an adjunct to existing therapies involving medication regimens.